Audible howling control systems and methods

Abstract

An audio system includes: a speaker; a microphone that generates a microphone signal based on sound output from the speaker; a mixer module configured to generate a mixed signal by mixing the microphone signal with an audio signal; a filter module configured to filter the mixed signal to produce a filtered signal and to apply the filtered signal to the speaker; and a detector module configured to determine a howling frequency in the microphone signal attributable to sound output from the speaker, where the filter module is configured to decrease a magnitude of the filtered signal at the howling frequency.

What is claimed is: 
     
       1. An audio system comprising:
 a speaker; 
 a microphone that generates a microphone signal based on sound output from the speaker; 
 a mixer module configured to generate a mixed signal by mixing the microphone signal with an audio signal; 
 a filter module configured to filter the mixed signal to produce a filtered signal and to apply the filtered signal to the speaker; and 
 a detector module configured to determine a howling frequency in the microphone signal attributable to sound output from the speaker, 
 wherein the filter module is configured to decrease a magnitude of the mixed signal at the howling frequency, and 
 wherein the detector module is configured to determine the howling frequency based on a comparison of the filtered signal applied to the speaker at a first time and the microphone signal at a second time that is after the first time. 
 
     
     
       2. The audio system of  claim 1  wherein the detector module includes a neural network trained to determine howling frequencies and the detector module determines the howling frequency using the neural network. 
     
     
       3. The audio system of  claim 2  wherein the neural network is a deep neural network. 
     
     
       4. The audio system of  claim 1  wherein the filter module includes a notch filter, and the filter module is configured to adjust a notch frequency range of the notch filter such that the howling frequency is within the notch frequency range. 
     
     
       5. The audio system of  claim 1  wherein the filter module includes a notch filter, and the filter module is configured to adjust a notch depth of the notch filter at the howling frequency. 
     
     
       6. An audio system comprising:
 a speaker; 
 a microphone that generates a microphone signal based on sound output from the speaker; 
 a mixer module configured to generate a mixed signal by mixing the microphone signal with an audio signal; 
 a filter module configured to filter the mixed signal to produce a filtered signal and to apply the filtered signal to the speaker; 
 a detector module configured to determine a howling frequency in the microphone signal attributable to sound output from the speaker, 
 wherein the filter module is configured to decrease a magnitude of the mixed signal at the howling frequency; and 
 a power spectral density (PSD) module configured to determine a PSD based on the microphone signal, 
 wherein the detector module is configured to determine the howling frequency based on the PSD. 
 
     
     
       7. The audio system of  claim 6  wherein the PSD module is further configured to determine a second PSD based on the filtered signal, and
 wherein the detector module is configured to determine the howling frequency further based on the second PSD. 
 
     
     
       8. The audio system of  claim 7  wherein the detector module includes a neural network trained to determine howling frequencies based on PSDs and the detector module determines the howling frequency using the neural network. 
     
     
       9. The audio system of  claim 8  wherein the neural network is a deep neural network. 
     
     
       10. A vehicle comprising:
 the audio system of  claim 1 ; and 
 a passenger cabin, 
 wherein the speaker outputs sound within the passenger cabin, and 
 wherein the microphone is disposed within the passenger cabin. 
 
     
     
       11. A method comprising:
 by a microphone, generating a microphone signal based on sound output from a speaker; 
 generating a mixed signal by mixing the microphone signal with an audio signal; 
 filtering the mixed signal to produce a filtered signal; 
 applying the filtered signal to the speaker; and 
 determining a howling frequency in the microphone signal attributable to sound output from the speaker, 
 wherein determining the howling frequency includes determining the howling frequency based on a comparison of the filtered signal applied to the speaker at a first time and the microphone signal at a second time that is after the first time; and 
 decreasing a magnitude of the mixed signal at the howling frequency. 
 
     
     
       12. The method of  claim 11  wherein determining the howling frequency includes determining the howling frequency using a neural network trained to determine howling frequencies. 
     
     
       13. The method of  claim 11  wherein the filtering includes applying a notch filter and adjusting a notch frequency range of the notch filter such that the howling frequency is within the notch frequency range. 
     
     
       14. The method of  claim 11  wherein the filtering includes applying a notch filter and adjusting a notch depth of the notch filter at the howling frequency. 
     
     
       15. The method of  claim 11  further comprising:
 determining a power spectral density (PSD) module based on the microphone signal; and 
 determining the howling frequency further based on the PSD. 
 
     
     
       16. The method of  claim 15  further comprising:
 determining a second PSD based on the filtered signal; and 
 determining the howling frequency further based on the second PSD. 
 
     
     
       17. The method of  claim 16  wherein determining the howling frequency includes determining the howling frequency using a neural network trained to determine howling frequencies based on PSDs. 
     
     
       18. The method of  claim 17  wherein the neural network is a deep neural network.